feat(content-ops): ship reviewed X composer experience seed

loopx/capabilities/content_ops/README.md

@@ -91,3 +91,88 @@ compatibility facade and delegates those packets without changing their output.
 The provider does not open a browser, read a timeline, or publish. A real
 browser session remains owner-controlled, and every external write still needs
 the exact account, body, media/link plan, source references, and stop condition.
+
+### Bundled experience before X preparation / 发 X 前的经验种子
+
+Every new Agent can read the versioned, public-safe
+[`x-composer-preflight-v1.json`](experiences/x-composer-preflight-v1.json)
+from either existing provider entry point, without a memory account or network:
+
+```bash
+loopx value-connectors source-map --connector social_browser_x --format json
+loopx value-connectors install-check --connector social_browser_x --format json
+```
+
+The JSON `operating_experience` field contains the full seed, version, package
+content digest, applicability, observations, procedure and memory initialization
+recipe. Markdown output exposes the compact lesson and points to JSON. The
+seed ships in the wheel, not only in the source tree. Read it before editing a
+composer, and revalidate the final body/card after every subsequent edit.
+
+新的 Agent 不需要继承旧会话，也不需要先有 memory provider，就能从以上入口读到
+完整经验种子。它属于 `content-ops` 的内置 `social_browser_x` provider；不新增
+capability、扩展、内核规则或第二套记忆存储。
+
+The seed includes a negative publication result: entering the main URL last
+restored its composer preview, but a published long post bound `card.rest_id`
+to the first supporting URL in its `note_tweet` URL entities instead. Input
+order, body order and published-card selection are distinct. Retain every
+required link and try the intended URL first in the actual body; this remains
+a mitigation **without a verified successful published readback**, not an X
+contract. Unknown target binding stays unknown. A rich-text `fill` can leave old paragraphs behind;
+scoped native select-all/clear/insert plus exact readback is the recovery path.
+Composer readiness and published-post verification are separate evidence. Do
+not remove requested links or publish a test merely to obtain a card.
+
+这里保存的是包含失败反例的操作经验：最后输入主链接只能改变编辑器预览，曾经
+在发布后仍绑定到第一个合作者链接。保留多个链接，把目标链接放到正文第一个
+URL 位置，是待发布后验证的缓解方案，不是已证实的修复。输入先后、正文顺序和
+发布端选卡必须分开。不得为验证卡片擅自发帖，也不得用删链接代替解决问题。
+
+### Initialize an Agent's own memory / 初始化自己的记忆
+
+Bundled discovery is implemented; automatic seed import is intentionally **not**
+performed. A host can initialize its own provider through the existing
+Reward Memory `scoped_feedback` adapter, after corpus-owner authorization and
+current-artifact review. Repository publication is not permission to write a
+user's provider, and another Agent's private corpus is not a shared seed store.
+
+1. Read `operating_experience` and deduplicate by the configured Goal/Agent scope
+   plus seed id, version and `content_digest`. Review newer revisions rather
+   than overwriting local corrections or reviving retired memories.
+2. Resolve the Agent's existing
+   [Reward Memory configuration](../reward_memory/README.md). Keep provider
+   credentials and scope references in its owner-local configuration.
+3. Prepare one `scoped_feedback_reward_memory_event_v0` using the seed's
+   `content_summary`, `target_class=procedural_experience`, and source kind
+   `reviewed_learning_card`. Use the seed id/version/digest as a stable source
+   reference; fill workspace/project/user/peer and surface from the configured
+   corpus. Set `requested_action_scopes=[]` and `raw_content_captured=false`.
+   The configured standing policy still reviews the event. Set
+   `current_artifact_verified=true` only after actual verification; package
+   presence alone does not satisfy it. Never copy a ready-made verification flag.
+4. Use the existing scoped import and require its exact provider readback:
+
+   ```bash
+   # event.json contains adapter, event and observed_at; preview first.
+   loopx reward-memory ingest-event --goal-id <goal> --agent-id <agent> \
+     --input event.json --format json
+   # Only under the corpus owner's existing write authorization:
+   loopx reward-memory ingest-event --goal-id <goal> --agent-id <agent> \
+     --input event.json --execute --format json
+   ```
+
+5. For subsequent automatic retrieval, use the existing
+   [`agent_workflow.turn_admission`](../agent_turn_recall/README.md) surface,
+   configured to the compatible corpus with automatic recall enabled. Name
+   actual X preparation in the selected Todo so its bounded query can retrieve
+   the relevant experience. Memory availability is not proof of recall, and
+   recall is not proof that the Agent applied or verified the lesson.
+
+经验种子读取、写入个人 memory、下一轮召回、当前操作验证，是四个不同结果。
+未配置或不可用时，直接参考仓库种子并如实报告，不冒称“已入库/已召回”。后续
+平台行为变化应由新的验证经验修订或淘汰旧记录；不要提升为全局强制偏好。
+
+No frontend/Lark configuration changes are introduced: these commands only
+read packaged public guidance and retain existing memory settings and gates.
+There are no new background imports, browser operations or external writes.

loopx/capabilities/content_ops/catalog_entry.py

@@ -22,6 +22,11 @@
     ),
     "entry_command": "loopx content-ops aggregate-packets --format json",
     "commands": [
+        {
+            "command": "loopx value-connectors source-map --connector social_browser_x --format json",
+            "purpose": "Before X preparation, read the provider's bundled operating experience and optional memory initialization recipe.",
+            "write_boundary": "packaged seed read only; no browser or memory-provider calls",
+        },
         {
             "command": "loopx content-ops exploration-plan --format json",
             "purpose": "Plan source lanes before reading connector material.",

loopx/capabilities/content_ops/experiences/x-composer-preflight-v1.json

@@ -0,0 +1,54 @@
+{
+  "schema_version": "content_ops_provider_experience_seed_v0",
+  "seed_id": "social_browser_x.composer_preflight",
+  "version": 1,
+  "observed_on": "2026-09-13",
+  "provider_id": "social_browser_x",
+  "target_class": "procedural_experience",
+  "authority": "advisory",
+  "title": "Preserve multiple X links while verifying the intended preview card",
+  "applicability": [
+    "Preparing or revising an X browser composer with multiple URLs",
+    "Handing an editable draft back for manual publication"
+  ],
+  "content_summary": "X composer previews can disagree with published cards. A long post bound its card to the first supporting URL despite a correct main-link preview. Preserve all links; try the intended URL first in the actual body, not merely last in input order. This mitigation is unproven without published readback. Native clear/insert and exact readback avoid rich-text duplication. Never publish to test without authorization.",
+  "observations": [
+    "A supporting PR or repository URL can replace the intended main-link card.",
+    "Entering the intended URL last restored a composer preview but did not preserve that selection after publication. This is a failed publication mitigation, not a success recipe.",
+    "In a published long post observed on 2026-09-10, card.rest_id matched the first supporting URL in note_tweet entities rather than the intended fourth URL; the card lacked renderable content.",
+    "Moving the main URL first later retained a supporting preview in the editor; no corresponding successful published readback was recorded. Body order and input order are distinct, and neither is a universal guarantee.",
+    "A rich-text fill operation can replace only part of an existing editor, leaving duplicated paragraphs."
+  ],
+  "procedure": [
+    "Read the current composer and preserve the owner's latest wording, links and account selection.",
+    "Record the intended card URL separately from supporting URLs; do not silently delete links.",
+    "Try placing the intended URL before all supporting URLs in the actual body, preferably near the opening. Record this as a mitigation pending published readback, not a proven fix. Merely typing it last is insufficient. Use scoped native editing and verify the editor is empty before inserting a replacement body.",
+    "Read back the complete final body. Wait for the card to resolve, then inspect its title, loaded image and target where exposed. Missing target evidence remains unknown; do not infer it from the body URL alone.",
+    "After any text or link edit, repeat card verification. If ambiguous, leave the draft unsent and report the gap.",
+    "Hand off at the requested manual-send boundary. After an authorized publication, inspect the actual post and card target. Where available, resolve card.rest_id through the long-post URL entities and check renderable card data; do not confuse note_tweet content with the truncated legacy body. A correct composer is preparation evidence only."
+  ],
+  "limits": [
+    "URL order is an observed mitigation, not a stable X API contract or machine-enforced policy.",
+    "Cached metadata, removed previews and platform changes can produce different results.",
+    "This seed is a public distilled observation, not a raw chat, screenshot, account record or permission grant."
+  ],
+  "memory_initialization": {
+    "mode": "caller_reviewed_opt_in",
+    "adapter": "scoped_feedback",
+    "source_kind": "reviewed_learning_card",
+    "recall_surface_id": "agent_workflow.turn_admission",
+    "identity_key": ["goal_id", "agent_id", "seed_id", "version", "content_digest"],
+    "query_hint": "X browser draft preparation: multi-link card selection, rich-text replacement and final readback",
+    "instructions": [
+      "Use the actor's explicitly configured corpus and provider; do not discover or reuse another Agent's private memory.",
+      "Review this seed against current artifacts and higher-priority instructions. Fill the existing scoped_feedback event with the configured scope, this content_summary and a seed/version/digest source reference.",
+      "Do not mark current_artifact_verified merely because the seed exists. Import as procedural_experience only after current verification through the existing candidate/review and ingest-event path.",
+      "Require the existing provider write and exact readback receipt before claiming initialization. Deduplicate the same scoped seed revision; review updates rather than overwriting newer local experience.",
+      "For automatic recall, configure the existing agent_workflow.turn_admission surface and include the actual X preparation work in the selected Todo. A configured provider alone does not connect a recall surface.",
+      "If memory is disabled or unavailable, read this bundled seed directly and report bundled guidance, not provider recall."
+    ],
+    "provider_calls_performed": false,
+    "automatic_import": false,
+    "grants_publish_authority": false
+  }
+}

loopx/capabilities/content_ops/social_browser_x.py

@@ -1,6 +1,9 @@
 from __future__ import annotations
 
+import hashlib
+import json
 import shutil
+from importlib.resources import files
 from typing import Any
 
 
@@ -12,13 +15,22 @@
 
 
 def build_social_browser_x_provider_packet() -> dict[str, Any]:
+    # Public packaged guidance, not a second memory store or an import side effect.
+    seed_bytes = (
+        files("loopx.capabilities.content_ops")
+        .joinpath("experiences/x-composer-preflight-v1.json")
+        .read_bytes()
+    )
+    experience = json.loads(seed_bytes)
+    experience["content_digest"] = "sha256:" + hashlib.sha256(seed_bytes).hexdigest()
     return {
         "ok": True,
         "schema_version": CONTENT_OPS_SOCIAL_BROWSER_X_PROVIDER_SCHEMA_VERSION,
         "connector_id": SOCIAL_BROWSER_X_CONNECTOR_ID,
         "outcome_capability_id": "content-ops",
         "provider_module": SOCIAL_BROWSER_X_PROVIDER_MODULE,
         "source_profile": {
+            "operating_experience": experience,
             "connector_id": SOCIAL_BROWSER_X_CONNECTOR_ID,
             "status": "profile_ready_when_browser_available",
             "route_type": "browser-backed public/social channel",
@@ -50,6 +62,7 @@ def build_social_browser_x_provider_packet() -> dict[str, Any]:
             ],
         },
         "install_check": {
+            "operating_experience": experience,
             "connector_id": SOCIAL_BROWSER_X_CONNECTOR_ID,
             "status": "ready" if shutil.which("ego-browser") else "needs_ego_browser",
             "install": [

loopx/capabilities/value_connectors/install_check.py

@@ -119,7 +119,7 @@
     }
 
 
 def render_value_connector_install_check_markdown(payload: dict[str, Any]) -> str:
     lines = ["# LoopX Value Connector Install Check", ""]
     for item in payload.get("checks") or []:
         if not isinstance(item, Mapping):
@@ -134,6 +134,14 @@
         )
         if item.get("write_gate"):
             lines.append(f"- write_gate: {item.get('write_gate')}")
+        experience = item.get("operating_experience")
+        if isinstance(experience, Mapping):
+            lines.append(
+                f"- bundled experience `{experience.get('seed_id')}` v{experience.get('version')}: {experience.get('content_summary')}"
+            )
+            lines.append(
+                "- Read operating_experience in the JSON packet before browser preparation; memory initialization remains caller-reviewed and opt-in."
+            )
         lines.extend(["", "Install / use:"])
         for command in item.get("install") or []:
             lines.append(f"- `{command}`")

loopx/capabilities/value_connectors/source_map.py

@@ -94,6 +94,7 @@ def _source_profile(
     maturity_hint: str,
     write_gate: str | None = None,
     stop_conditions: list[str] | None = None,
+    operating_experience: Mapping[str, Any] | None = None,
 ) -> dict[str, Any]:
     return {
         "schema_version": VALUE_CONNECTOR_SOURCE_PROFILE_SCHEMA_VERSION,
@@ -121,6 +122,11 @@ def _source_profile(
             "requested action would perform an external write without an audit gate",
         ],
         **_outcome_provider_binding(connector_id),
+        **(
+            {"operating_experience": dict(operating_experience)}
+            if operating_experience is not None
+            else {}
+        ),
     }
 
 
@@ -426,6 +432,14 @@ def render_value_connector_source_map_markdown(payload: dict[str, Any]) -> str:
             )
         if profile.get("write_gate"):
             lines.append(f"- write_gate: {profile.get('write_gate')}")
+        experience = profile.get("operating_experience")
+        if isinstance(experience, Mapping):
+            lines.append(
+                f"- bundled experience `{experience.get('seed_id')}` v{experience.get('version')}: {experience.get('content_summary')}"
+            )
+            lines.append(
+                "- Read operating_experience in the JSON packet for applicability and opt-in memory initialization; bundled guidance is not provider recall or publish authority."
+            )
         lines.append("")
     action_gated = payload.get("action_gated_profiles")
     if isinstance(action_gated, list) and action_gated:

pyproject.toml

@@ -59,7 +59,7 @@ include = ["loopx*"]
 "loopx.control_plane.turn_driver" = ["*.ts"]
 "loopx.control_plane.work_items" = ["*.ts"]
 "*" = ["README.md", "README.zh-CN.md", "docs/*.md", "docs/protocols/*.md"]
-"loopx.capabilities.content_ops" = ["templates/*.json"]
+"loopx.capabilities.content_ops" = ["templates/*.json", "experiences/*.json"]
 "loopx.capabilities.manager_context" = ["skills/loopx-manager/SKILL.md"]
 "loopx.extensions.lark" = ["extension.toml", "README.md", "docs/*.md"]
 "loopx.extensions.openviking_periodic_report" = ["extension.toml"]